filmov
tv
UFS Webinar Series: Improving weather forecast skill and rainfall climatology of FV3GFS using ML
Показать описание
Chris Bretherton, Senior Director, Vulcan Climate Modeling and Professor of Atmospheric Sciences, University of Washington
Abstract: Vulcan Climate Modeling (a small philanthropically-supported group in Seattle) and NOAA/GFDL are collaborating on a pilot project to use machine learning to develop a skillful corrective parameterization for a full-complexity global atmospheric model that helps it evolve more like a reference data set, which could be a reanalysis or a finer-grid global model. We have applied this approach to climate-oriented versions of FV3GFS with 100-200 km grids. Encouragingly, it significantly improves their 0-7 day weather forecasts and time-mean precipitation distribution, both in present and SST-perturbed climates.
Abstract: Vulcan Climate Modeling (a small philanthropically-supported group in Seattle) and NOAA/GFDL are collaborating on a pilot project to use machine learning to develop a skillful corrective parameterization for a full-complexity global atmospheric model that helps it evolve more like a reference data set, which could be a reanalysis or a finer-grid global model. We have applied this approach to climate-oriented versions of FV3GFS with 100-200 km grids. Encouragingly, it significantly improves their 0-7 day weather forecasts and time-mean precipitation distribution, both in present and SST-perturbed climates.
0:57:50
0:57:50
0:53:42
1:03:05
1:00:28
0:51:47
1:04:27
0:57:16
0:47:11
0:58:17
0:57:43
1:05:52
0:37:41
0:56:09
0:54:48
0:53:48
0:51:49
1:04:24
0:59:10
0:51:17
1:05:53
0:58:46
1:01:45
1:04:24